A novel application of the lorenz plot method for ultra-short-term evaluation of autonomic nervous function

Heart rate variability (HRV) analysis is widely used to assess autonomic nervous function (ANF); however, conventional frequency-domain indices require recording durations that may limit their applicability in settings requiring rapid or repeated measurements. This study investigated whether measurement duration could be reduced by estimating high-frequency power (HF), an index of parasympathetic modulation, from ultra-short-term (UST) recording windows using a Lorenz plot (LP)-based estimation model.

Electrocardiographic data were recorded during a 2-min resting, eyes-closed condition. HF was calculated exclusively from the full 120-s recording using frequency-domain analysis and served as the reference outcome. LP indices were computed from progressively shorter R-R interval (RRI) windows (120-5 s) and, together with age, were used to construct regression models to estimate the 120-s HF reference value. No frequency-domain analysis was performed on the shortened windows.

The LP-based estimation model demonstrated robust performance across window lengths. Even when LP indices were computed from as little as 5 s of data, the model retained acceptable predictive accuracy for estimating the HF reference value derived from the 120-s recording (adjusted R² = 0.67).

Parasympathetic modulation reflected in conventional 120-s frequency-domain HF analysis can be estimated with practical accuracy from UST recordings as short as 5 s using the LP-based estimation model. The proposed model does not confer interchangeability with standard frequency-domain HF indices across time scales. Rather, it offers a practical marker for rapid estimation of the 120-s frequency-domain HF reference value when longer recordings are impractical.